Whether the undetermined gonad primordium develops into an ovary or testis is controlled by genes on the gonosomes and presumably by one or more autosomal loci. Within species of Xiphophorus XX individuals occasionally develop into functional males and XY individuals may become females. Such exceptional individuals are observed in increased frequency in interspecific hybrids. These cases of atypical sex determination (ASD) are apparently controlled by autosomal factors. The proposed experiments are designed to analyze how many factors or loci are involved in the YY female and XX male phenomena within species and in interspecific hybrids, and to determine to which autosomal biochemical markers or pigment genes the factors for ASD are linked. XX males (from an XX female-XX male line) of X. nigensis will be outcrossed to a different species, X. pygmaeus, with no history of XX males, F1 females will then be backcrossed to the original males. From the frequency of XX males in the backcross generation it is possible to calculate the number of autosomal factors involved. This process will be repeated for two or four more generations to determine whether the results are reproducible. The progeny of the 1st backcross generation will be examined electrophoretically to determine whether male differentiation is correlated with any of the 80 protein-coding loci known to be polymorphic or variant in the genus. Interspecific crosses are much more likely to detect such linkage than intraspecific crosses, because F1 species hybrids are often variant at more than 30 autosomally controlled biochemical and pigmentary markers. The chances that each of the 23 pairs of autosomes is marked are pretty good. If no linkage can be detected, the crosses will be repeated substituting a third species for the outcross. Similar crosses will be used to study XX males of X. milleri and YY females of X. maculatus. The possibility that the autosomal factors and the sex-determining genes on the gonosomes are species-specific will be examined in maculatus x milleri hybrids. All sex chromosomes carry unique marker genes. If the autosomal factor of maculatus interact atypically with Xmi, XX males can be expected among the XmaXmi but not XmaXma progeny in the 1st backcross to maculatus. This experimental system can be changed one species at a time. Linkage analysis will be as above. The proposed experiments provide a model to explain ASD in man which is known to be genetic in nature, but the mechanism is not understood. In Xiphophorus different kinds of ASD apparently result from specific interaction of certain sex chromosomes with specific autosomal factors.